In the third generation communication scheme, typically such as IMT-2000 (International Mobile Telecommunications-2000), in particular, a faster and higher-capacity downlink is sought, and for example, the information transmission rate of higher than or equal to 2 Mbps is achieved by using the frequency band of 5 MHz. IMT-2000 adopts the single carrier Wideband-Code Division Multiple Access (W-CDMA) scheme. Alternately, some scheme called High Speed Downlink Packet Access (HSDPA) may be employed. In HSDPA, the Adaptive Modulation and channel Coding (AMC) scheme, MAC layer packet Automatic Repeat Request (ARQ) scheme, fast packet scheduling and others are employed to achieve higher transmission rates and higher quality. For example, AMC is described in non-patent document 1. ARQ is described in non-patent document 2.
FIG. 1 is a schematic view for explaining the AMC scheme. Assuming that transmission power from a base station is fixed, in general, a terminal 11 closer to a base station 10 can receive signals with greater power than a terminal 12 far from the base station 10. Hence, since it is estimated that the terminal 11 has a better channel state, a greater modulation level and a higher coding rate are adopted. On the other hand, the terminal 12 receives signals with less power than the terminal 11. Thus, since it is estimated that the terminal 12 does not have a good channel state, a smaller modulation level and a lower coding rate are adopted.
FIG. 2 shows an exemplary combination of different modulation schemes (modulation level) and different channel coding rates. In the illustrated table, the rightmost column represents relative bit rates in the case of the bit rate being “1” under the modulation scheme M of “QPSK” and the channel coding rate R of “1/3”. For example, if M=“QPSK” and R=“1/2”, the bit rate of ×1.5 is obtained. In general, there is tendency that the higher bit rate is, the less reliability is. More specifically, combinations between different modulation schemes and the coding rates and different amounts indicative of channel states are predefined in a listing table, and the modulation schemes and others are changed depending on the channel state if needed. The amount indicative of the channel state is managed as Channel Quality Indicator (CQI), which is typically SIR (Signal to Interference power Ratio) and SINR of a received signal.
FIG. 3 is a schematic view for explaining the ARQ (more accurately, hybrid ARQ). The hybrid ARQ scheme is a technique derived from a combination of the ARQ scheme of requesting retransmission of packets depending on results of error detection (CRC: Cyclic Redundancy Check) and some error correction coding scheme (also referred to as channel coding) for error correction. As illustrated, a CRC bit is added to a transmission data sequence S1), and the resulting signal is sent after completion of error correction encoding (S2). In response to receipt of the signal, error correction decoding (also referred to as “channel decoding”) is carried out (S3), and error detection is carried out (S4). If some error is detected, retransmission of the packet is requested to the transmitting side (S5). As illustrated in FIG. 4, there are several methods for such retransmission.
In an exemplary method illustrated in FIG. 4A, packet P1 is sent from the transmitting side to the receiving side. If some error is detected at the receiving side, the packet P1 is discarded and then the retransmission is requested. In response to the retransmission request, the transmitting side resends the same packet (represented as “P2”) as the packet P1.
In an exemplary method illustrated in FIG. 4 B, packet P1 is sent from the transmitting side to the receiving side. If some error is detected at the receiving side, the receiving side keeps the packet P1 without discarding it. In response to the retransmission request, the transmitting side resends the same packet (represented as “P2”) as the packet P1. Then, the receiving side generates packet P3 by combining the previously received packet with the currently received packet. Since the packet P3 corresponds to one transmitted with double the power of packet P1, the demodulation accuracy is improved.
Also in an exemplary method illustrated in FIG. 4C, packet P1 is sent from the transmitting side to the receiving side. If some error is detected at the receiving side, the receiving side keeps the packet P1 without discarding it. In response to the retransmission request, the transmitting side sends redundancy data derived by performing certain operations on the packet P1 as packet P2. For example, assume that a sequence of packets such as “P1, P1′, P1″, . . . ” has been derived by encoding the packet P1. The derived sequence may differ depending on adopted coding algorithms. In the illustrated example, in response to receipt of a retransmission request, the transmitting side sends P1′ as packet P2. The receiving side generates packet P3 by combining the previously received packet with the currently received packet. Since the packet P3 has increased redundancy, the demodulation accuracy will be improved. For example, assuming that the coding rate of the packet P1 is equal to “1/2”, the coding rate of the packet P3 becomes equal to “1/4”, thereby resulting in improved reliability. Note that the receiving side must already know some information as to what coding algorithm is adopted, what redundancy data are sent (also referred to as “puncture pattern”), and others.
Fast packet scheduling scheme is a technique intended to improve frequency utilization efficiency in downlink. In a mobile communication environment, the channel state between a mobile station (user) and a base station varies over time as illustrated in FIG. 5. In this case, even though transmission of a large amount of data to a user with poor channel state is attempted, it is hard to improve the throughput. On the other hand, the higher throughput would be achieved for a user with a good channel state. From such a viewpoint, it is possible to improve the frequency utilization efficiency by determining whether the channel state is good for each user and assigning a shared data packet in favor of the user with the better channel state.
FIG. 5 is a schematic diagram for explaining the fast packet scheduling scheme. As illustrated, a shared data packet is assigned to a user with the better channel state (a user associated with greater SINR) in each time slot.
Non-patent document 1: T. Ue, S. Aampei, N. Morinaga and K. Hamaguchi, “Symbol Rate and Modulation Level-Controlled Adaptive Modulation/TDMA/TDD System for High-Bit-Rate Wireless Data Transmission”, IEEE Trans. VT, pp. 1134-1147, vol. 47, No. 4, November 1998
Non-patent document 2: S. Lin, Costello, Jr. and M. Miller, “Automatic-Repeat-Request Error Control Schemes”, IEEE Communication Magazine, vol. 12, No. 12, pp. 5-17, December 1984